Abstract:
PURPOSE: An ontology information browsing method, an ontology information browsing device thereof, and a storage medium including a program implementing an ontology information browser thereof display a large amount of ontology information in a simple way and make it easier for a user to select which piece of ontology information to be displayed. CONSTITUTION: A transceiver unit (310) communicates with a network and receives a request for ontology information. Based on the received ontology information, a browser operation unit (320) launches an ontology information browser. A display unit (340) displays the search results of ontology information the browser identifies based on the received ontology information. The ontology information browser places the information requested by a user at the edge of the display unit's screen and with the requested information as a center, related pieces of ontology information are placed on several concentric circles with different distances from the center. [Reference numerals] (310) Transceiver unit; (320) Browser operation unit; (330) Storage unit; (340) Display unit
Abstract:
PURPOSE: A graph data-based instance route search system and a method thereof are provided to efficiently search for correlations by applying instance information and relation information between classes to a parallel graph database. CONSTITUTION: A peer server information database(160) stores information about a class route managed by peer servers. A graph database(110) stores class information about keywords and relation information. If two or more keywords are inputted by a user, a master server(100) extracts a class about the keywords from the graph database and confirms the peer servers from the peer server information database in order to transmit a route search command including the class to the peer servers. [Reference numerals] (100) Master server; (110) Graph DB; (120) Communication unit; (130) User interface unit; (140) Class route generating unit; (150) Route search request processing unit; (160) Peer server information DB
Abstract:
PURPOSE: A data distribution system and a method thereof are provided to minimize network IO time and join operations between nodes in a distribution parallel system, thereby shortening response time of the whole system. CONSTITUTION: Data nodes(200a-200c) store data. A management node(100) confirms a type pattern by analyzing the data. The management node determines a data node which stores the data based on state information of data nodes which the type pattern is set and distributes the data to the data node. When the data includes a plurality of type patterns, the management node is assigned to one data node. [Reference numerals] (100) Management node; (200a) Data node 1; (200b) Data node 2; (200c) Data node n; (AA) Data
Abstract:
PURPOSE: An outputting device for technology association relation information of an institution, an outputting method thereof, and an application storage medium thereof are provided to allow associated institutions to obtain association information at a time. CONSTITUTION: A system comprises a storage unit(210), an association relation extraction module(220), and an outputting unit(230). The storage unit stores technology relation information, institution technology association information, and top/bottom technology information corresponding to a technology range. The association relation extraction module processes a search based on a keyword inputted by a user and provides the technology relation information, the association relation extraction module, and the top/bottom technology information corresponding to the technology range. Institution association relation information is generated by using data which is inputted or selected by the user. The association relation extraction module provides information, which is generated by delivering the institution association relation information to the outputting unit, to the user. [Reference numerals] (210) Storage unit; (220) Association relation extraction module; (230) Outputting unit
Abstract:
PURPOSE: A scheme matching device and a method thereof are provided to accurately determine a response relation between terminal nodes in a scheme by measuring similarity through location and matching relations between nodes. CONSTITUTION: A scheme parser(110) generates a hierarchical structure by parsing a target scheme. A language similarity measuring unit(120) measures a language similarity value between terminal nodes in the hierarchical structure. A terminal node matching unit(130) selects a matched terminal node and determines a response relation of the terminal nodes by using the language similarity value. If two or more terminal nodes are selected, a structure similarity measuring unit(140) obtains a neighboring node similarity value and a path similarity value of the terminal nodes. The terminal node matching unit selects a matching terminal node having the biggest similarity value and determines the response relation. [Reference numerals] (100) Scheme matching device; (110) Scheme parser; (120) Language similarity measuring unit; (130) Terminal node matching unit; (140) Structure similarity measuring unit; (160) Matching node information DB
Abstract:
PURPOSE: An entity identifying result search system and a method thereof are provided to use attribute information of multiplex ontology included in Linked Data, thereby gradually identifying various types of entities. CONSTITUTION: A multiplex ontology database(300) stores attribute information about entities as ontology. When a query s inputted, an entity identifying result search system(100) obtains the identifying result of the query from a triple storage device, compares the identifying result with the identifying result of the query, and determines need for entity identification. If the entity identification is necessary, the entity identifying result search system obtains and compares the attribute information of the identification target entity from the multiplex ontology database, distinguishes the entities, and provides the result information based on the entity identification. The triple storage device stores the entity identifying result of the entity in a triple type. [Reference numerals] (100) Entity identifying result search system; (300a,300n) Ontology database
Abstract:
본 발명은 동적 임계값이 적용된 문서 브라우징 장치 및 방법에 관한 것으로, 입력되거나 저장된 문서에 대하여 각 문서 사이의 유사도를 관리하는 문서 관리 모듈, 문서 관리 모듈을 검색하여 기준 문서에 대하여 설정된 임계값 이상의 유사도를 가지는 문서를 검색하는 유사 문서 검색 모듈 및 검색된 문서들 중 어느 하나를 선택하고 선택된 문서로부터 설정된 임계값 이상의 유사도를 가지는 문서들을 연속적으로 검색하여 브라우징 연결 구성을 생성하는 브라우징 연결 구성 모듈을 포함한다.
Abstract:
검색 결과가 보장된 질의어 유형별 자동완성 시스템 및 방법 및 프로그램을 기록한 기록매체가 개시된다. 이를 위하여 질의어를 입력하는 검색 시스템에서 검색 결과가 보장되는 질의어 검색 결과만을 자동완성하여 제시하는 것으로, 특히, 입력된 질의어에 대하여 검색결과가 존재하는 질의어 목록과 유형을 제시하고 문서정보의 추가, 삭제 등에 실시간으로 대응할 수 있기 때문에 검색 결과의 신뢰도를 높이고, 검색의 실패를 방지하며 검색을 신속하게 하고, 유형별 정보를 그룹화하여 개선상태로 제공하는 효과가 있다. 자동완성, 검색, 문서정보, 색인어, 빈도수, 질의어, 데이터베이스, 서버, 주제어
Abstract:
본 발명은 입력되는 추론규칙을 적어도 하나 이상의 유형으로 분류하는 추론규칙 유형 분류수단; 상기 분류된 추론규칙의 유형에 근거하여 네트워크를 생성하는 네트워크 생성수단; 그리고 소정의 트리플을, 상기 생성된 네트워크에 적용하여 확장된 트리플을 도출하는 네트워크 실행수단을 포함하는 메모리 및 DBMS 기반의 하이브리드 Rete 추론 시스템 및 방법이 개시된다. 추론규칙, RDF 트리플, 메모리 추론방식(Rete 추론방식), DBMS 추론방식